Soluble markers of inflammation are associated with Framingham scores in HIV-infected patients on suppressive antiretroviral therapy

Journal of Infection (2011) 63, 382e390

www.elsevierhealth.com/journals/jinf

Soluble markers of inflammation are associated with Framingham scores in HIV-infected patients on suppressive antiretroviral therapy n-Fulgencio a, Jose Medrano b, Norma Rallo  n b, Marı´a Guzma Amaya Echeverria-Urabayen c, Jose Miguel Benito b, Clara Restrepo b, a   nica Garcı´a-Alvarez Mo , Eugenia Vispo b, Jesus San Roman d, b nchez-Piedra , Vicente Soriano b, Salvador Resino a,* Carlos Sa a

Laboratory of Molecular Epidemiology of Infectious Diseases, Centro Nacional de Microbiologı´a, Instituto de Salud Carlos III (Campus Majadahonda), Carretera Majadahonda-Pozuelo, Km 2.2, 28220 Majadahonda, Madrid, Spain b Infectious Diseases Department, Hospital Carlos III, Madrid, Spain c Service of Neurology, Fundacion Jime´nez Dı´az, Madrid, Spain d Department of Medicine, School of Medicine, University Rey Juan Carlos, Madrid, Spain Accepted 8 August 2011 Available online 12 August 2011

KEYWORDS Cardiovascular risk; Framingham score; Endothelial dysfunction; Chronic immune activation; Biomarkers; AIDS; Antiretroviral therapy

Summary Objective: To evaluate the association between biomarkers of inflammation and endothelial dysfunction and Framingham scores (FS) for risk of coronary heart disease (FSCHD), stroke (FS-Stroke) or any cardiovascular event (FS-CVE) in HIV-infected on suppressive highly active antiretroviral therapy (HAART). Methods: A cross-sectional study was conducted in 73 HIV-infected patients and 23 healthy controls. Inflammatory molecules and endothelial dysfunction markers were measured using a multiplex immunoassay (plasminogen activator inhibitor type 1 (PAI-1), soluble TNF receptor type 1 (sTNF-R1), soluble CD40 ligand (sCD40L), soluble E-selectin (sE-selectin), soluble P-selectin (sP-selectin), soluble intercellular adhesion molecules (sICAM-1) and soluble vascular cell adhesion molecule (sVCAM-1). Outcome variables were FS-CHD
10%, FS-Stroke
5% and FS-CVE
10%. Results: Significant differences (p < 0.05) were found comparing controls and HIV patients for PAI-1 (5.4 vs. 13.5 ng/dL), sTNF-R1 (0.85 vs. 1.09 ng/dL), sICAM-1 (529 vs. 858 ng/dL), sEselectin (73.7 vs. 120 ng/dL), sP-selectin (676 vs. 1511 ng/dL) sCD40L (76 vs. 307 ng/dL), FSCHD (4% vs. 7.8% L), FS-Stroke (2% vs. 2.8%) and FS-CVE (5% vs. 11%). In HIV-infected patients, the adjusted logistic regression analysis revealed that sTNF-R1 levels were significantly

* Corresponding author. Tel.: þ34 918 223 266; fax: þ34 915 097 946. E-mail address: [email protected] (S. Resino). 0163-4453/$36 ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2011.08.006

Soluble biomarkers of cardiovascular risk in HIV

383

associated with increased FS-CHD>10% (OR: 11.51 (95% CI: 1.14; 115.84); p Z 0.038) and FSCVE (OR: 12.41 (95% CI: 1.25; 123.23); p Z 0.031). Conclusions: HIV-infected patients show higher levels of soluble inflammatory and endothelial dysfunction markers than controls and have a two-fold increased FS of presenting coronary heart disease, stroke or cardiovascular events at 10 years. Furthermore, sTNF-R1 displayed the best association with FS of coronary heart disease and any cardiovascular event in our patients. ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved.

Introduction

patients on suppressive HAART taking as reference a group of HIV-negative individuals matched by age and gender.

The advent of highly active antiretroviral therapy (HAART) has made HIV infection a chronic manageable disease in the western world, where mortality due to AIDS has dramatically declined,1 but other causes of death, such as cardiovascular complications, have emerged in this population.2,3 HIV infection itself and antiretroviral drugs may further contribute to the increased risk of cardiovascular disease characteristically seen in HIV-infected individuals.4,5 A pivotal role for inflammation has been postulated to be the link between chronic HIV infection and cardiovascular disease,6 given that immune activation persists even when virus replication is suppressed and CD4 counts relatively well preserved using antiretroviral therapy.7 Thus, chronic viral infections can trigger immune activation and inflammation, whose persistence may result in vascular damage and ultimately atherosclerosis,8,9 but cardiovascular risk among HIV-infected individuals could be reduced through early initiation of antiretroviral therapy, before CD4 T-cell counts are depressed.10 On the other hand, some antiretroviral drugs are associated with dislipidemia and other metabolic abnormalities which, in turn, may enhance the cardiovascular risk.4,6,9 Estimating the risk of developing clinical cardiovascular events with Framingham score (FS) has been used to drive preventive and therapeutic interventions in the general population,11 and to predict the absolute risk of developing major cardiovascular events at 5 and 10 years.12,13 These Framingham risk scores are calculated using the traditional risk factors for cardiovascular disease.11 Among these factors, hypertension, insulin resistance, diabetes mellitus type 2, dyslipidemia, smoking, and lack of exercise are overrepresented in the HIV population and efforts to properly manage these conditions are encouraged to reduce the incidence of cardiovascular events in this population.6,14 Moreover, Pavia consensus suggested that FS may be used as cardiovascular risk marker in HIV-infected patients on HAART15; and it may represent adequate surrogate endpoints to prevent vascular events and premature aging.16,17 In this context, these cardiovascular risk scores were considered clinically relevant among HIV patients in several recent reports.18,19 One of the links between metabolic syndrome and cardiovascular risk is the chronic inflammation which is associated with endothelial activation,20 and both are evident in HIV-infected individuals,21 supporting a potential role of inflammation in endothelial dysfunction and cardiovascular disease in HIV infection.22 Over the last decade, several soluble biomarkers of inflammation have demonstrated to be predictors of future cardiovascular events in HIV-infected individuals.23 The aim of our study was to evaluate the association of biomarkers of inflammation and endothelial dysfunction with FS in HIV-infected

Patients and methods Study design A cross-sectional study was carried out on 73 HIV-infected patients receiving HAART on regular follow-up at a reference HIV clinic located in Madrid. Subjects were seen from December 2008 to June 2009 and had to fulfill the following inclusion criteria: a) unchanged HAART regimen for at least the last 3 months, b) undetectable plasma HIV-RNA, c) CD4 counts above 200 cells/mL, d) negative serum hepatitis B surface antigen and plasma HCV-RNA, and e) absence of any known inflammatory condition other than HIV. A control group of healthy HIV-negative individuals matched for age and gender was selected. This group was represented by professionals working at the same institution who voluntarily agreed to act as controls for several studies. The local Institutional Ethics Committee approved the study and all subjects gave written informed consent.

Clinical and laboratory data Medical records and questionnaires were used to obtain clinical and epidemiological data. Body weight, height, waist and hip measurements and body mass index (BMI) were calculated (kg/m2) and recorded for each individual. Subjects were asked to complete an electronic questionnaire collecting data on lifestyle, drinking and smoking habits, and medical history. The information from all questionnaires was stored in a database for subsequent analyses. For the purpose of this study, HIV-infected patients were considered to have lipodistrophy when this diagnosis was recorded in the clinical records. After filling the questionnaire, a blood sample was obtained from each participant. In controls, the blood test also included serologies for HIV, HBV and HCV, as well as the complete lipid profile. Blood specimens were coded and plasma samples immediately stored and frozen at 70  C for further assays. Plasma HIV-RNA was measured using the third-generation branched DNA assay (Quantiplex version 3.0; Siemens, Barcelona, Spain), which displays a lower detection limit of 50 copies/mL. CD4þ T cells were measured using the Coulter TetraOne reagent and an EPICS XL flow cytometer (Coulter Instruments, Madrid, Spain).

Soluble biomarkers A multiplex suspension bead array immunoassay was performed using the FlowCytomix kit (Comboplex Bender

n-Fulgencio et al. M. Guzma

384 MedSystems, Vienna, Austria) and the FACSCalibur flow cytometer (BD FACSCalibur, BD Biosciences, CA). Manufacturer’s instructions were followed for measuring plasminogen activator inhibitor type 1 (PAI-1), soluble TNF receptor type 1 (sTNF-R1), soluble CD40 ligand (sCD40L), soluble E-selectin (sE-selectin), soluble P-selectin (sP-selectin), soluble intercellular adhesion molecules (sICAM-1) and soluble vascular cell adhesion molecule (sVCAM-1) in frozen plasma specimens.

Framingham scores The FS are validated tools that predict the absolute risk of developing major cardiovascular events at 5 and 10 years.12,13,24 Depending on specific clinical outcomes considered, coronary heart disease (CHD), stroke or both, three distinct FS risk modalities can be distinguished. Each of them include distinct variables: a) Coronary heart disease (FS-CHD): age, total blood cholesterol, high density lipoprotein (HDL) cholesterol, systolic blood pressure (SBP), diabetes, and smoking; b) Stroke (FS-Stroke): age, diabetes, smoking, blood pressure, antihypertensive therapy, prior cardiovascular disease, atrial fibrillation, and left ventricular hypertrophy; and c) Any Cardiovascular event (FS-CVE), which is a composite score including FSCHD plus FS-Stroke. To achieve the greatest predictive power, gender-specific functions were applied. Moreover, given that the relative weight of age and arrhythmia are higher in the FS-Stroke than in FS-CHD, appropriate corrections were made. All Framingham scores were calculated for each eligible patient by inputting the requested variables into a web-based calculator on the West Hertfordshire Cardiology website: http://www.westhertshospitals.nhs.uk/whc/archive/calculators/Cardiac%20Risk %20Assessments.xls.

Statistical analysis Categorical data and proportions were analyzed using the chi-squared test or Fisher’s exact test, as required. The ManneWhitney U test was used to compare data between independent groups. To evaluate the association between inflammatory (PAI-1, sTNF-R1) and endothelial dysfunction markers (sCD40L, sE-selectin, sP-selectin, sICAM-1, and sVCAM-1), linear regression models were tested. Since these variables had a non-linear distribution, a log10 transformation was made before entering analyses. To avoid confounding between the associations of different soluble markers, all tests were adjusted by CD4 nadir, CD4 count, BMI, lipodystrophy, and time on HAART. Logistic regression analyses were performed to evaluate the association of inflammatory and endothelial dysfunction markers with the highest cardiovascular risk estimates (FS-CHD
10%, FSStroke
5% and FS-CVE
10%). These regression analyses were also adjusted by CD4 nadir, current CD4 count, BMI, lipodystrophy, and time on HAART. All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) v15.0 (SPSS Inc., Chicago, IL). All p-values were two-tailed and statistical significance was defined for p-values <0.05.

Results Characteristics of the study population The main characteristics of HIV-infected patients and controls are recorded in Table 1. All HIV-infected persons had undetectable viremia, were mostly men who had acquired the infection through sex and had an average length of infection of 14 years. Their mean age was 46 years-old and all were whites. Median nadir and current CD4 counts were 299 and 659 cells/mL, respectively. All patients had been treated with NRTI (mean time exposure, 221 months), 79.5% with NNRTI (mean time exposure, 35.5 months) and 76.7% with PI (mean time exposure, 42.1 months). Lipodystrophy was recorded in 14 patients (19.2%). No significant differences were found comparing distinct demographic variables between groups. BMI was similar in both populations. Although 8 HIV-infected patients received therapy for hypertension, significantly higher values of systolic and diastolic blood pressure were found in the HIV group than in controls. Any subject had a previous diagnosis of cardiovascular disease. Only one HIV-infected patient was receiving therapy for type II diabetes. Around half of individuals in both groups were smokers (47.8% of controls and 56.2% of patients). Triglycerides were significantly higher in HIV patients than in controls (132.0 vs. 73.5 mg/dL, p Z 0.001). Finally, a trend was noticed for HIV patients who had lower values of HDL-cholesterol than controls (45.5 vs. 41.0 mg/dL, p Z 0.069). None of the subjects was receiving lipid-lowering therapy.

Clinical characteristics and inflammation We found a significant association among several clinical characteristics (glucose, triglycerides, systolic blood pressure, and BMI levels) and high PAI-1 levels; and, current smoker and systolic blood pressure levels and high sTNF-R1 levels (Table 2).

Inflammation and endothelial dysfunction biomarkers Overall, levels of soluble markers were significantly higher in HIV-infected patients than in controls except for sVCAM1. Considering only inflammatory markers, significant differences were found comparing controls and HIV patients for PAI-1 (5.4 vs. 13.5 ng/dL, p < 0.001) and sTNF-R1 (0.85 vs. 1.09 ng/dL, p Z 0.042). For endothelial dysfunction markers, differences were found for sICAM-1 (529 vs. 858 ng/dL, p Z 0.006), sE-selectin (73.7 vs. 120 ng/dL, p Z 0.009), sP-selectin (676 vs. 1511 ng/dL, p < 0.001) and sCD40L (76 vs. 307 ng/dL, p Z 0.014). As previously mentioned, no significant differences were found for sVCAM-1 when comparing controls and HIV patients (1534 vs. 1482 ng/dL, p Z 0.533) (Fig. 1). Linear regression analyses were performed in HIVinfected patients, which revealed a positive association between inflammatory and endothelial dysfunction markers (Table 3). In unadjusted linear regression analyses, PAI-1 levels were significantly associated with sICAM-1, sVCAM-

Soluble biomarkers of cardiovascular risk in HIV Table 1

385

Main epidemiological and clinical characteristics of the study population.

Parameter

Healthy controls

HIV-1 patients

p

No. Malea Age (years)b HIV-1 infection HIV acquired by IVDUa Years of infectionb Nadir CD4þ T cells/mLb CD4þ T cells/mLb Antiretroviral therapy NRTIa Time on NRTI (months)b NNRTIa Time on NRTI (months)b PIa Time on PI (months)b Lipodystrophy BMI (kg/m2)b Atrial fibrillation Arterial hypertension SBP (mm Hg)b DBP (mm Hg)b History of diabetes History of cardiovascular disease Left ventricular hypertrophy Current smokera Biochemical parametersb Plasma glucose (mmol/L) Total cholesterol (mg/dL) HDL (mg/dL) LDL (mg/dL) Triglycerides (mg/dL)

23 14 (60.9%) 43 (36; 48)

73 58 (79.5%) 46 (40; 50)

0.064 0.169

e e e e

16 (21.9%) 14 (9.5; 18) 299 (228; 490) 659 (444; 913)

e e e e

e e e e e e e 24.5 (22.9; 26.5) 0 0 120 (106; 128) 71 (67; 81) 0 0 0 11 (47.8%)

70 (95.9%) 221 (103; 317.7) 58 (79.5%) 35.5 (16; 72.6) 56 (76.7%) 42.1 (17.1; 92.1) 14(19.2) 24.5 (22.1; 26.6) 0 8 134 (124; 147) 82 (74.5; 90) 1 0 0 41 (56.2%)

e e e e e e e 0.707 e e <0.001 0.001 e e e 0.322

97 (84; 103.2) 181 (160.2; 200) 45.5 (39.7; 51.7) 123 (99; 140) 73.5 (56.2; 110.5)

97 (93; 105) 179 (156; 198) 41 (34; 49) 104 (84; 123) 132 (93.5; 183)

0.126 0.261 0.069 0.038 0.001

Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; IVDU, intravenous drug users; HDL, high density lipoprotein; LDL, low density lipoprotein; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor. a Absolute number (percentage). b Median (percentiles 25 and 75). In bold, values with statistical significance.

Table 2 Association of epidemiological and clinical markers of cardiovascular risk with inflammation markers in HIV-infected patients.

Gender (male) Current smoker Lipodystrophy Log10 age (years) Log10 glucose (mmol/L) Log10 cholesterol (mg/dL) Log10 triglycerides (mg/dL) Log10 HDL (mg/dL) Log10 LDL (mg/dL) Log10 SBP (mmHg) Log10 SBP (mmHg) Log10 BMI

Log10 PAI-1 (ng/mL)

Log10 sTNF-R1 (ng/mL)

0.01 0.07 0.01 0.01 1.21 0.49 0.26 0.36 0.04 1.35 0.22 1.10

0.07 0.15 0.10 0.13 0.61 0.44 0.07 0.06 0.04 1.37 0.66 0.11

           

0.05 0.05 0.06 0.31 0.48* 0.37 0.10* 0.22 0.25 0.43** 0.36 0.28**

           

0.07 0.06* 0.08 0.41 0.68 0.51 0.15 0.31 0.33 0.39* 0.48 0.42

Values expressed as regression lineal coefficient (b) and standard error of mean (s.e.m.). Theses values were adjusted by CD4 nadir, CD4þ count, and time on HAART. Statistical significant differences: * (p < 0.05); ** (p < 0.01). Abbreviations: PAI-1, plasminogen activator inhibitor 1; sTNFR-1, soluble tumor necrosis factor receptor 1; HDL, high density lipoprotein; LDL, low density lipoprotein; SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index.

n-Fulgencio et al. M. Guzma

386 Control

HIV-1

A

B **

1800

16

1600

1000

**

ng//mL

ng/mL

**

800 600

1.6

10

1.4

8 6

**

200 sICAM-1

sVCAM-1

sE-selectin

0

sCD40L

1.2 1 0.8 0.4

2 sP-selectin

*

0.6

4

400

0

1.8

12

ng/mL

1400 1200

**

14

0.2 PAI-1

0

sTNF-R1

Figure 1 Serum levels (mean and standard error) of inflammatory and endothelial dysfunction markers in healthy controls and HIV-infected patients. Statistical significant differences: * (p < 0.05); ** (p < 0.01). Abbreviations: PAI-1, plasminogen activator inhibitor 1; sTNF-R1, soluble tumor necrosis factor receptor 1; MCP-1, Monocyte chemotactic protein-1; sICAM-1, soluble intercellular adhesion molecule-1; sVCAM-1, soluble vascular cell adhesion molecule-1; sE-selectin, soluble endothelial selectin; sPselectin, soluble platelet selectin; sCD40L, soluble CD40 ligand.

Cardiovascular risk Overall, HIV-infected patients had a two-fold increased risk of developing cardiovascular events than controls. In the control group, the risk of developing coronary heart disease, stroke or any cardiovascular event estimated by FS-CHD, FS-Stroke and FS-CVE was 4.7%, 1.9% and 6.5%, respectively. These figures in HIV-infected patients on suppressive HAART were 8.6%, 3.2% and 11.9%, respectively. Differences for each score were highly significant in all cases (Fig. 2). In HIV-infected patients, unadjusted logistic regression analyses revealed that PAI-1 and sTNF-R1 levels were positively associated with increased risk of coronary heart disease and any cardiovascular events but stroke (Table 4). However, adjusted logistic regression analyses revealed that only sTNF-R1 levels were associated with increased risk of coronary heart disease and any cardiovascular event (Table 4). In contrast with inflammatory markers, none of the endothelial dysfunction biomarkers (sICAM-1, sVCAM1, sE-selectin, sP-selectin, and sCD40L) was significantly associated with any of the Framingham scores of cardiovascular risk (data not shown).

Discussion Cardiovascular events have emerged as one of the leading causes of shortened life expectancy in HIV-infected patients receiving HAART.2,3 In our study, HIV-infected patients displayed higher levels of almost all soluble markers of inflammation and endothelial dysfunction than healthy controls, despite HIV-RNA suppression under

HAART. These findings and the fact that levels of soluble inflammatory molecules were associated to those of soluble endothelial dysfunction markers, support that there is a pathogenic link between residual HIV replication, inflammation, endothelial dysfunction and atherosclerosis.8,25 Moreover, inflammation may also be due to non-specific factors (not HIV replication only) such as hyperlipemia, insulin resistance, and smoking.26 In addition, HIV patients on HAART who develop cardiovascular disease usually have multiple risk factors for this condition such as significantly greater prevalence of smoking, insulin resistance/impaired glucose intolerance, hypertension, and higher cholesterol and triglyceride levels.6,14 In our study, we found positive association among a number of clinical markers of smoking, cardiovascular risk (glucose, triglycerides, systolic blood pressure, and BMI) and inflammatory markers (sTNF-R1 and PAI-1). This association could indicate a link between Control

HIV-1

14

Framingham Score (FS. %)

1, sE-selectin, sP-selectin and sCD40L levels. Similarly, sTNF-R1 levels were strongly associated with sICAM-1, sVCAM-1, sE-selectin, sP-selectin and sCD40L levels. When regression analyses were adjusted by CD4 nadir, current CD4 count, BMI, lipodistrophy and time on HAART, similar significant associations were found between inflammatory and endothelial dysfunction markers, except for PAI-1 with both sICAM-1 and sVCAM-1 (Table 3).

**

12 **

10 8 6

**

4 2 0

CHD

Stroke

CVE

Figure 2 Mean Framingham scores in healthy controls and HIV-infected patients. Statistical significant differences: * (p < 0.05); ** (p < 0.01). Abbreviations: FS-CHD, Framingham score for coronary heart disease risk; FS-Stroke, Framingham score for stroke risk; FS-CVE, Framingham score for any cardiovascular event risk.

Soluble biomarkers of cardiovascular risk in HIV

387

Table 3 Summary of lineal regression assessing associations between levels (log10 ng/mL) of inflammatory and endothelial dysfunction markers in HIV-infected patients.

Univariate PAI-1 sTNF-R1 Multivariatea PAI-1 sTNF-R1

sICAM1

sVCAM1

sE-selectin

sP-selectin

sCD40L

0.20  0.10* 0.45  0.10**

0.12  0.05* 0.23  0.06**

0.30  0.08** 0.46  0.09**

0.50  0.07** 0.34  0.09**

1.02  0.17** 0.84  0.21**

0.21  0.19 0.34  0.12**

0.02  0.08 0.16  0.05**

0.31  0.15* 0.38  0.09**

0.25  0.09* 0.18  0.06**

1.06  0.31** 0.80  0.20**

Values expressed as regression lineal coefficient (b) and standard mean error. Statistical significant differences: * (p < 0.05); ** (p < 0.01). Abbreviations: PAI-1, plasminogen activator inhibitor 1; sTNFR-1, soluble tumor necrosis factor receptor 1; sE-selectin, soluble endothelial selectin; sP-selectin, soluble platelet selectin; sICAM-1, soluble intercellular adhesion molecule-1; sVCAM-1, soluble vascular cell adhesion molecule-1; sCD40L, soluble CD40 ligand. a Values were adjusted by CD4 nadir, CD4þ count, body mass index, lipodystrophy, and time on HAART.

inflammation and metabolic disturbance,27 which have been associated to increased cardiovascular risk.28,29 Furthermore, smoking and high SBP are two classical cardiovascular risk factors which are also very prevalent in the HIV population.6 In our study, we observed that HIV patients on HAART had an activation of the TNF-a system, measured through the plasma concentrations of sTNF-R1 and PAI-1. Although TNF-R1 induces adhesion molecule expression (i.e., ICAM-1 and VCAM-1) on endothelial cells,30 the pathogenic meaning of sTNF-R1 in atherosclerosis remains unclear. In HIV negative persons, sTNF-R1 has been suggested as a marker of atherosclerosis, given its relationship with carotid lesions especially in patients aged <70 years31; has been associated with an increased risk of coronary heart disease in women but not in men32 and mortality and heart failure after myocardial infarction,33 and are currently being assessed in secondary stroke prevention.34 However, in prior publications in HIV-infected patients, sTNF-R1 was not associated with carotid intima-media thickness.22 It is noteworthy that levels of sTNF-R1 in our study were significantly associated with both high levels of endothelial dysfunction molecules and FS. In this regard, measurement of sTNF-R1 could act as surrogate of risk of coronary heart disease and cardiovascular events in HIV-infected patients on

suppressive HAART and periodic monitoring of this biomarker might identify those at increased risk in whom preventive measures might be stressed. It must be noted, however, that the cross-sectional design of our study and the estimation of cardiovascular risk with FS instead of real observation of clinical events over time, both limit the strength of our conclusions. Furthermore, the association of PAI-1 with FS was weaker than for TNF-R1. PAI-1, a major inhibitor of fibrinolysis, is elevated in metabolic syndrome and lipodystrophy, and has proved an independent risk factor for cardiovascular disease.35,36 In addition, the increased values of PAI-1 have been associated to deregulation of TNF, with decreased fibrinolysis and increased coagulability in HIV patients on HAART37; representing an additional risk factor for cardiovascular disease in these patients.38 Moreover, these two inflammatory markers (sTNF-R1 and PAI-1) appear to be independent of HIV replication because they are high in our HIV patients on HAART with undetectable viral load. We also found positive association among a traditional markers of cardiovascular risk (glucose, triglycerides, systolic blood pressure, and BMI) and inflammatory markers (sTNF-R1 and PAI-1) (data no shown). These associations could indicate a link between inflammation and metabolic disturbance,27 which have been associated

Table 4 Summary of logistic regression examining associations between inflammatory markers (log10 ng/mL) and Framingham scores of cardiovascular risk (10-years risk) in HIV-infected patients. FS-CHD risk
10% Univariate PAI1 sTNF-R1 Multivariatea PAI1 sTNF-R1

FS-Stroke risk
5%

FS-CVE risk
10%

OR (95%CI)

p

OR (95%CI)

p

OR (95%CI)

p

13.92 (1.60; 118.94) 6.70 (1.10; 40.70)

0.017 0.039

16.15 (0.83; 312.86) 6.25 (0.72; 53.94)

0.066 0.095

15.46 (2.27; 105.26) 10.40 (1.62; 66.91)

0.005 0.013

12.59 (0.41; 380.24) 11.51 (1.14; 115.84)

0.145 0.038

10.72 (0.20; 574.22) 5.07 (0.51; 50.62)

0.243 0.166

13.86 (0.59; 324.33) 12.41 (1.25; 123.23)

0.102 0.031

Values expressed as regression odds ratio (OR) and 95% confidence interval (95%CI). In bold, values with statistical significance. Abbreviations: FS-CHD, Framingham score for coronary heart disease risk; FS-Stroke, Framingham score for stroke risk; FS-CVE, Framingham score for cardiovascular event risk; PAI-1, plasminogen activator inhibitor 1; sTNF-R1, soluble tumor necrosis factor receptor 1. a Values were adjusted by CD4 nadir, CD4 count, body mass index, lipodystrophy, and time on HAART.

n-Fulgencio et al. M. Guzma

388 to increased cardiovascular risk.28,29 Thus, the excessive release of lipids, cytokines and other pro-inflammatory products from adipose tissue could induce insulin resistance, endothelial dysfunction, abnormal fibrinolysis, and formation of atherosclerotic plaques along the vascular tree.35,36,39,40 Although we used controls and most baseline variables did not differ significantly in both groups, the higher prevalence of hypertension and diabetes in HIV-infected patients compared to controls might have contributed to explain the higher risk of coronary heart disease, stroke and any cardiovascular event in HIV patients compared to controls. The recognition of an association between inflammatory biomarkers and coronary heart disease but not with stroke must be interpreted cautiously. We examined a relatively young population, in which the risk of stroke was very low, which might in part have limited the recognition of any association between soluble markers and stroke. Our findings are consistent with the poor of correlation between FS-CHD and FS-Stroke seen in other studies that also examined relatively young populations.12,13,24,41 Another interesting finding in our study was that endothelial dysfunction markers (sICAM-1, sVCAM-1, sE-selectin, sP-selectin, and sCD40L), despite being significantly elevated in HIV-infected patients, were not associated with the risk of cardiovascular events. A poor correlation between the concentration of soluble endothelial molecules in the bloodstream and the extent of atherosclerotic vascular damage might explain this observation.42 Finally, we briefly highlight the most important limitations of our study, which may have an impact our findings: a) the study has a cross-sectional design in a small cohort of HIV-infected patients without a uniform HAART; b) the control group is relatively small, and although differences in age and gender are not significant, the sample size is different between the study groups. Besides, potential contributions by confounding from unmeasured factors when comparing HIV-infected patients with working professionals (control group) might compromise the validity of the study, even if we excluded control subjects with chronic inflammatory diseases and heavy hereditary burden of cardiovascular disease; c) Due to the absence of prospective follow-up, there was no data of cardiovascular events or an indirect method of measuring the development of subclinical atherosclerosis such as the measurement of carotid intima-medial thickness (IMT) and brachial artery flow-mediated dilatation (FMD). This is especially relevant since both internal and common carotid-artery intimamedia thickness have recently been identified as independent predictors of cardiovascular events in HIV un-infected patients.43 d) The use of risk score thresholds as cardiovascular endpoints, although simplifying the estimation of the outcome, may be of uncertain validity when attempting to draw inferences about the pathogenesis of cardiovascular disease. In our study, we did not evaluate the high sensitivity (hs) C-reactive protein (hsCRP) and hs troponin levels, two biomarkers which may replace the large panel of determinations investigated in the study. Instead, we found that sTNF-R1 and PAI-1 were associated with endothelial dysfunction markers and FS of CHD and CVE risk in our study. Specifically, TNF-R1 could be considered an adequate

prognostic marker of cardiovascular risk among HIV patients on HAART and undetectable viral load because this biomarker had a strong association with FS in adjusted regression analysis. In summary, our results suggest that HIV-infected patients receiving HAART have higher levels of soluble markers of inflammation and endothelial dysfunction than controls and have a two-fold increased FS of presenting coronary heart disease, stroke or cardiovascular events at 10 years. Further studies examining longitudinally the prognostic value of sTNF-R1 as biomarker of cardiovascular disease are warranted.

Authors’ contributions MGF and JM had primary responsibility for protocol development, participated in the design of the study, performed the statistical analysis, and contributed to the writing of the manuscript. MGF, NR, CR, JMB, and MGA collecting and recording sample and data, and they carried out the immunoassays and contributed to the writing of the manuscript. JM, JAEU, EV, JSR, JMB, and VSF carried out patient screening, and contributed to the writing of the manuscript. SR conceived of the study, and participated in its design and coordination. All authors read and approved the final manuscript.

Acknowledgments Potential conflicts of interest: none for all authors. Financial support: This work has been partially funded by grants from Instituto de Salud Carlos III (PI08/0738); Fundacion Investigacion y Educacion en SIDA (F-IES), Agencia Lain Entralgo, Red de Investigacion en SIDA (RIS, project ISCIII-RETIC RD06/006) and the European NEAT network.  n-Fulgencio, Mo nica Garcı´a-Alvarez, Marı´a Guzma and Jose Medrano are supported by grants from Instituto de Salud Carlos III (CM09/00031, CM08/00101, and CM09/ 00010, respectively).

References 1. Mocroft A, Ledergerber B, Katlama C, Kirk O, Reiss P, D’Arminio Monforte A, et al. Decline in the AIDS and death rates in the EuroSIDA study: an observational study. Lancet 2003 Jul 5; 362(9377):22e9. 2. Lang S, Mary-Krause M, Cotte L, Gilquin J, Partisani M, Simon A, et al. Increased risk of myocardial infarction in HIV-infected patients in France, relative to the general population. AIDS 2010 May 15;24(8):1228e30. 3. Lewden C, May T, Rosenthal E, Burty C, Bonnet F, Costagliola D, et al. Changes in causes of death among adults infected by HIV between 2000 and 2005: the "Mortalite 2000 and 2005" surveys (ANRS EN19 and Mortavic). J Acquir Immune Defic Syndr 2008 Aug 15;48(5):590e8. 4. Lorenz MW, Stephan C, Harmjanz A, Staszewski S, Buehler A, Bickel M, et al. Both long-term HIV infection and highly active antiretroviral therapy are independent risk factors for early carotid atherosclerosis. Atherosclerosis 2008 Feb;196(2):720e6. 5. Worm SW, Sabin C, Weber R, Reiss P, El-Sadr W, Dabis F, et al. Risk of myocardial infarction in patients with HIV infection exposed to specific individual antiretroviral drugs from the 3

Soluble biomarkers of cardiovascular risk in HIV

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

major drug classes: the data collection on adverse events of anti-HIV drugs (D: A:D) study. J Infect Dis 2010 Feb 1;201(3): 318e30. Aberg JA. Cardiovascular complications in HIV management: past, present, and future. J Acquir Immune Defic Syndr 2009 Jan 1;50(1):54e64. Hunt PW, Martin JN, Sinclair E, Bredt B, Hagos E, Lampiris H, et al. T cell activation is associated with lower CD4þ T cell gains in human immunodeficiency virus-infected patients with sustained viral suppression during antiretroviral therapy. J Infect Dis 2003 May 15;187(10):1534e43. Baker J, Quick H, Hullsiek KH, Tracy R, Duprez D, Henry K, et al. Interleukin-6 and d-dimer levels are associated with vascular dysfunction in patients with untreated HIV infection. HIV Med 2010 Oct 1;11(9):608e9. Monsuez JJ, Charniot JC, Escaut L, Teicher E, Wyplosz B, Couzigou C, et al. HIV-associated vascular diseases: structural and functional changes, clinical implications. Int J Cardiol 2009 Apr 17;133(3):293e306. Ho JE, Deeks SG, Hecht FM, Xie Y, Schnell A, Martin JN, et al. Initiation of antiretroviral therapy at higher nadir CD4þ T-cell counts is associated with reduced arterial stiffness in HIVinfected individuals. AIDS 2010 Jul 31;24(12):1897e905. Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, Minhas R, Sheikh A, et al. Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2. BMJ 2008 Jun 28;336(7659):1475e82. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998 May 12;97(18):1837e47. D’Agostino RB, Wolf PA, Belanger AJ, Kannel WB. Stroke risk profile: adjustment for antihypertensive medication. The Framingham Study. Stroke 1994 Jan;25(1):40e3. Lekakis J, Ikonomidis I, Palios J, Tsiodras S, Karatzis E, Poulakou G, et al. Association of highly active antiretroviral therapy with increased arterial stiffness in patients infected with human immunodeficiency virus. Am J Hypertens 2009 Aug;22(8):828e34. Volberding PA, Murphy RL, Barbaro G, Barbarini G, Bruno R, Cirelli A, et al. The pavia consensus statement. AIDS 2003 Apr;17(Suppl 1):S170e9. Rossi R, Nuzzo A, Guaraldi G, Orlando G, Squillace N, Ligabue G, et al. The role of the Framingham risk score to predict the presence of subclinical coronary atherosclerosis in patients with HIV infection. J Acquir Immune Defic Syndr 2009 Oct 1;52(2):303e4. Law MG, Friis-Moller N, El-Sadr WM, Weber R, Reiss P, D’Arminio Monforte A, et al. The use of the Framingham equation to predict myocardial infarctions in HIV-infected patients: comparison with observed events in the D: A:D study. HIV Med 2006 May;7(4):218e30. Edwards-Jackson N, Kerr S, Tieu H, Ananworanich J, Hammer S, Ruxrungtham K, et al. Cardiovascular risk assessment in persons with HIV infection in the developing world: comparing three risk equations in a cohort of HIV-infected Thais. HIV Med; 2011 Mar 6. Falcone EL, Mangili A, Skinner S, Alam A, Polak JF, Wanke CA. Framingham risk score and early markers of atherosclerosis in a cohort of adults infected with HIV. Antivir Ther 2011;16(1):1e8. Tousoulis D, Charakida M, Stefanadis C. Endothelial function and inflammation in coronary artery disease. Heart 2006 Apr; 92(4):441e4. Aukrust P, Muller F, Lien E, Nordoy I, Liabakk NB, Kvale D, et al. Tumor necrosis factor (TNF) system levels in human immunodeficiency virus-infected patients during highly active antiretroviral therapy: persistent TNF activation is associated with virologic and immunologic treatment failure. J Infect Dis 1999 Jan;179(1):74e82.

389 22. Ross AC, Rizk N, O’Riordan MA, Dogra V, El-Bejjani D, Storer N, et al. Relationship between inflammatory markers, endothelial activation markers, and carotid intima-media thickness in HIVinfected patients receiving antiretroviral therapy. Clin Infect Dis 2009 Oct 1;49(7):1119e27. 23. Baker JV, Duprez D. Biomarkers and HIV-associated cardiovascular disease. Curr Opin HIV AIDS 2010 Nov;5(6):511e6. 24. Wolf PA, D’Agostino RB, Belanger AJ, Kannel WB. Probability of stroke: a risk profile from the Framingham study. Stroke 1991 Mar;22(3):312e8. 25. Francisci D, Giannini S, Baldelli F, Leone M, Belfiori B, Guglielmini G, et al. HIV type 1 infection, and not short-term HAART, induces endothelial dysfunction. AIDS 2009 Mar 13; 23(5):589e96. 26. Calza L, Manfredi R, Pocaterra D, Chiodo F. Risk of premature atherosclerosis and ischemic heart disease associated with HIV infection and antiretroviral therapy. J Infect 2008 Jul;57(1): 16e32. 27. Currier JS. Cardiovascular risk associated with HIV therapy. J Acquir Immune Defic Syndr 2002 Sep 1;31(Suppl. 1):S16e23. discussion S4eS5. 28. Blankenberg S, Rupprecht HJ, Bickel C, Peetz D, Hafner G, Tiret L, et al. Circulating cell adhesion molecules and death in patients with coronary artery disease. Circulation 2001 Sep 18;104(12):1336e42. 29. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000 Mar 23; 342(12):836e43. 30. Zhang L, Peppel K, Sivashanmugam P, Orman ES, Brian L, Exum ST, et al. Expression of tumor necrosis factor receptor1 in arterial wall cells promotes atherosclerosis. Arterioscler Thromb Vasc Biol 2007 May;27(5):1087e94. 31. Elkind MS, Cheng J, Boden-Albala B, Rundek T, Thomas J, Chen H, et al. Tumor necrosis factor receptor levels are associated with carotid atherosclerosis. Stroke 2002 Jan;33(1): 31e7. 32. Pai JK, Pischon T, Ma J, Manson JE, Hankinson SE, Joshipura K, et al. Inflammatory markers and the risk of coronary heart disease in men and women. N Engl J Med 2004 Dec 16;351(25): 2599e610. 33. Valgimigli M, Ceconi C, Malagutti P, Merli E, Soukhomovskaia O, Francolini G, et al. Tumor necrosis factor-alpha receptor 1 is a major predictor of mortality and new-onset heart failure in patients with acute myocardial infarction: the cytokineactivation and long-term prognosis in myocardial infarction (C-ALPHA) study. Circulation 2005 Feb 22;111(7):863e70. 34. Elkind MS, Luna JM, Coffey CS, McClure LA, Liu KM, Spitalnik S, et al. The levels of inflammatory markers in the treatment of stroke study (LIMITS): inflammatory biomarkers as risk predictors after lacunar stroke. Int J Stroke 2010 Apr; 5(2):117e25. 35. Alessi MC, Peiretti F, Morange P, Henry M, Nalbone G, JuhanVague I. Production of plasminogen activator inhibitor 1 by human adipose tissue: possible link between visceral fat accumulation and vascular disease. Diabetes 1997 May;46(5):860e7. 36. Krause JC, Toye MP, Stechenberg BW, Reiter EO, Allen HF. HIVassociated lipodystrophy in children. Pediatr Endocrinol Rev 2005 Sep;3(1):45e51. 37. He G, Andersen O, Haugaard SB, Lihn AS, Pedersen SB, Madsbad S, et al. Plasminogen activator inhibitor type 1 (PAI1) in plasma and adipose tissue in HIV-associated lipodystrophy syndrome. Implications of adipokines. Eur J Clin Invest 2005 Sep;35(9):583e90. 38. Koppel K, Bratt G, Schulman S, Bylund H, Sandstrom E. Hypofibrinolytic state in HIV-1-infected patients treated with protease inhibitor-containing highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2002 Apr 15;29(5):441e9.

390 39. Hansson GK, Libby P. The immune response in atherosclerosis: a double-edged sword. Nat Rev Immunol 2006 Jul;6(7): 508e19. 40. Hansson GK. Atherosclerosis e an immune disease: the Anitschkov Lecture 2007. Atherosclerosis 2009 Jan;202(1):2e10. 41. Meseguer E, Labreuche J, Durdilly C, Echeverria A, Lavallee PC, Ducrocq G, et al. Prevalence of embolic signals in acute coronary syndromes. Stroke 2010 Feb;41(2):261e6.

n-Fulgencio et al. M. Guzma 42. Muller O, Hamilos M, Bartunek J, Ulrichts H, Mangiacapra F, Holz JB, et al. Relation of endothelial function to residual platelet reactivity after clopidogrel in patients with stable angina pectoris undergoing percutaneous coronary intervention. Am J Cardiol 2010 Feb 1;105(3):333e8. 43. Polak JF, Pencina MJ, Pencina KM, O’Donnell CJ, Wolf PA, D’Agostino Sr RB. Carotid-wall intima-media thickness and cardiovascular events. N Engl J Med 2011 Jul 21;365(3):213e21.